Pick an option on each card and press Done; your answer goes straight
back to the session. The raiser's pick is marked. "I'm not sure" is a real answer.
1
In plain terms. Real customer applications at situsgroup properties have been missing their 'application needs review' emails to the property manager for weeks because the background worker that sends them was running 25-day-old code — should someone dig through that window and follow up with the property managers who were missed, or is moving on with the fix now enough?
The full question, as raised
The Fargate renewal-worker (propflow-renewal-worker-prod) was running commit 9f9682063b (2026-09-05) in prod, confirmed via the task's PROPFLOW_GIT_SHA env var and live CloudWatch logs showing the pre-fix logCritical call site (src/lib/temporal/activities/pm-action-reminder.ts, context=PmActionReminderWorkflow). That's 25 days and ~2 fix commits (dd25ad568b sandbox-org skip, 3f39d1e0c3 Sentry dedup) stale, despite the ECR repo showing dozens of fresh image pushes today and a same-day deploy-pipeline fix (#14289) already merged. The application_review PM-action-reminder cadence silently failed to notify PMs at real customer properties (appfolio-situsgroup-property-8/13/42/44/45/51), not just the Fairhaven TEST sandbox properties the alert named. I triggered a forced redeploy (gh workflow run deploy-renewal-worker.yml -f env=prod -f force=true, run 36805823220) which should bring the worker current, but I can't determine how far back the staleness goes (worker image history / ECR push log is my only receipt) or whether PMs at those 6 customer properties need a manual follow-up on applications that never got reminded.
Why, and what was tried
The session raising this recommends: Audit the window, follow up with PMs Real leasing applications sat with nobody reminded to review them — that's revenue-adjacent and customer-facing, and backfilling/notifying is outside what Smith should decide or do alone (no outbound comms without an explicit ask). Tried first, unsuccessfully: Traced the alert from the Sentry issue through the trigger code, the sandbox-org gate, the appfolio-sync Lambda bundle (confirmed current), then to the ECS renewal-worker task definition's PROPFLOW_GIT_SHA env var and live CloudWatch logs, which proved the worker is running 25-day-stale code. Forced a redeploy myself (reversible, documented workflow_dispatch force input) but stopped short of auditing historical impact or contacting customers, since that needs a human call.